Competitive intelligence work for business development is not a smaller version of the landscape work done for a brand team. The question is different, the deadline is different, and the consequence of being wrong is a single large irreversible decision rather than a quarter of suboptimal messaging.
BD asks a different question
A brand team asks what the competitive situation is. BD asks something narrower and harder: if we sign this, what will this asset be competing against when it is capable of launching, and is the differentiation in the seller's deck still there at that point?
That reframing does most of the work. It rules out a large amount of conventional landscape content as irrelevant, and it makes a small number of things decisive:
- The competitive set at the projected launch window, which is a pipeline question rather than a marketed-products question
- Whether the asset's claimed advantage is on a dimension that will still matter by then
- What the likely comparator will be in a registrational trial run under those future conditions — which drives development cost, not just commercial upside
- Whether anyone else is running at the same target, and how far ahead they are
Standard of care at launch, not at signing
This is the single most common way a deal case goes wrong, and it is structural rather than careless. An asset in-licensed at Phase II may be five or more years from launch. Benchmarking it against currently approved therapy flatters it, because the products that will actually define the standard of care at launch are, today, in Phase III and therefore easy to leave out of a comparison of "approved options".
The correct comparison is against the expected standard of care in the launch window, which requires taking a view on assets that have not read out. That view will be uncertain. It should be stated with its uncertainty rather than omitted for being uncertain — omitting it is not neutrality, it is an assumption that nothing changes, which is the one assumption guaranteed to be false.
The comparator set is the deliverable
Before any analysis, write down the specific cells the asset would compete in: indication, line of therapy, biomarker or molecular subset, and monotherapy versus combination. Then list every asset — approved or in development — that would occupy those cells at the projected launch window.
That list is the diligence. Nearly every substantive question is answerable once it exists, and nearly none are answerable before.
The granularity point matters more in some areas than others; in oncology it is the whole problem, for the reasons set out in why oncology CI is a different job.
The deadline is the real constraint
Competitive diligence usually gets two to four weeks, sometimes less, inside a process someone else is running. That has a consequence most CI functions underrate: diligence capability is mostly determined before the deal appears.
A team that already maintains an indication-level landscape in its areas of interest can produce a defensible competitive view inside the window. A team that has to build the landscape from scratch cannot, and what happens instead is that the deal case gets built on the seller's framing with some sanity checks bolted on. The remedy is not faster analysis during diligence; it is maintained coverage of the therapeutic areas the company actually does deals in.
Reading the seller's framing
A well-prepared seller's materials are accurate and selectively framed at the same time. Things worth checking as a matter of routine:
- How the competitive set was drawn. A narrow definition of the market makes any asset look differentiated. Redraw it yourself before comparing.
- Which comparator the efficacy claim is against. Cross-trial comparison against a historical control is not evidence of superiority over current therapy, however it is presented.
- Whether the population is the commercial population. Impressive results in a selected subset raise the question of whether the registrational trial can enrol it, and whether the label would cover enough patients to matter.
- What is absent. A competitor programme missing from the seller's landscape is worth more attention than anything in it.
- The development plan's realism — if the comparator in the planned registrational trial will be obsolete by the time the trial reads out, the timeline and the cost are both wrong.
The constraints you have to work inside
Diligence work sits under confidentiality agreements, and frequently under clean-team arrangements that limit who inside the acquirer may see what. Two practical implications for CI: the competitive analysis usually has to be built from sources that are public in any case, which is a discipline rather than a limitation; and the existence of the process is itself confidential, so the ordinary primary-research route is unavailable or has to be run without any framing that would signal interest in the target.
That last point is worth taking seriously. Commissioning expert interviews about a specific asset during a live process is a disclosure risk, and the ethical constraints on elicitation do not relax because a deal is involved — see primary versus secondary intelligence for where those lines sit.
What the output should look like. Not a landscape deck. A short memo that states the comparator set, the expected standard of care at launch, the specific claims the asset would need to win on, an explicit judgement on whether it can, and the two or three findings that would change the answer if they turned out differently. Deal teams read that. They do not read forty slides, and a forty-slide deck usually indicates the analyst did not reach a conclusion. Where the asset is oncology and the comparator set has to be built at indication level under a deadline, analyst-led coverage of that indication is the thing that makes the window achievable at all.
After the deal
The diligence view is the first version of the asset's competitive case, not a one-off artefact. Carrying it forward — the comparator set, the assumptions, the specific things that would have to be watched — gives the development team a baseline to revise against and makes the eventual pre-launch work considerably cheaper. Most organisations discard it, and rebuild the same analysis three years later from nothing.